Objective Five

Douglas Hagedorn and Nicole Timoshenko

3 - Research Approach: Objective Five

Evaluate the research model in terms of its practical effectiveness and relative value for the assessment of Alvar Conservation activities on Manitoulin Island.

To evaluate the quality of the research model developed in this study, the strengths and weaknesses inherent to its design and implementation must be compared and contrasted objectively. In this manner, areas for improvement can be noted and facets that bolster the study's value can be highlighted. The discussion below seeks to explore the strengths and weaknesses revealed in this study and suggest improvements that could be made for future work done in this topic area.

Strengths of the Model:

One fundamental strength of this project can be found in the ease with which the MCE framework used can be modified to test and evaluate different scenarios where varying levels of bias or weighting particular criteria are desired. If the model developed for this research were to be used as a planning tool, flexibility in this sense would likely prove to be highly useful as various conditions and disturbance sensitivity levels could be examined before significant investments were made for more comprehensive, long-term feasibility studies. In this sense, our research model can potentially be used for both scoping and analysis purposes.

Another primary strength of this research model can be found in the ability to integrate multiple evaluation factors within a framework that provides a singular holistic output. The land parcel evaluation mechanism used allowed for a comprehensive assessment of multiple conservation opportunities on the island according to 6 unique factors in a way that would prove to be difficult if the assessment was performed independently for each of the six factors chosen without any sort of MCE component. This project framework also combines many aspects of several independent publications into a single source. Given that research efforts concerning Manitoulin Island have been somewhat limited to date, a project such as this that synthesizes and builds upon existing literature should be of considerable value

A final note on the strength of this model can be made concerning the purchase of the Quary Bay Nature Reserve property in 1999. This 266 hectare property is spread along 18 kilometers of the islands southwestern coastline, in the location predicted to be the most optimum parcel for conservation by our study and serves as welcome validation of the strength of this GIS model even though details about this reserve were very sparsely documented in literature concerning Manitoulin Island, not expressed in the data used, and did not come to light until the final stages of this study.

Weaknesses of the Model:

The weaknesses of this model are largely related to issues of data reliability and the lack of inclusion of other potential factors that could be highly significant. Some limitations were associated with the age, reliability and precision of data used, since the most recent land coverage data published by the Ontario Ministry of Natural Resources (OMNR) in nearly ten years old. The geography of Manitoulin Island may have experienced a fair degree of land use change since that time. Additionally, the meta-data provided within the OMNR Land Coverage dataset indicates that the classification accuracy from original remotely sensed images ranges from 85%-96%. This varied degree of accuracy is mainly due to limited precision in the classification formulas, and varies among land use classes (Ministry of Natural Resources Metadata). It is likely that the land coverage depicted in this data has introduced some inaccuracies into the evaluation process – especially considering that some land cover in this data set was classified as unknown. In this regard, improvements that could be made to the implementation of this model could involve obtaining or producing more recent and/or higher quality land coverage data. However, as obtaining such data may be highly time consuming or capital intensive, the costs and benefits of such an upgrade would have to be carefully considered.

Within the model, certain design limitations exist as well. The lack of availability of property ownership data as a conservation suitability indicator made its inclusion into the evaluation process impossible. The land tracts identified as most suitable for conservation in our research outputs may already have an alternate land use, or may not be available for purchase. The incorporation of up-to-date land ownership as an evaluation criteria would have provided this project with a greatly increased level of relevance to the landscape context.

Our model is also limited by the method used for evaluation parcel delineation. The contiguous land parcels that were created for this study were only included in the analysis if they were larger than 300 hectares. Consequently many of the small off shore islands have been left out of our analysis, although they may contain land that is suitable for Alvar conservation initiatives. The evaluation parcels were also created using a 150 m buffer between their perimeter and the closest roads so that facilities and disturbances located in close proximity to the road network would not be included in land being evaluated for conservation potential. However, the true extent to which this 150m buffer excludes all near road activities could not be determined, so it is possible that an alternate buffer distance would have been more appropriate.

A final point of weakness for this project can be found in the possibility that some or all of the GPS coordinates provided for the Alvar sites listed in the Ontario Ministry of Natural Resources Natural Heritage Information Catalogue (2007) and The Alvars of Ontario (2000), may be incorrect as a result the practice of scrambling GPS signals that was employed for security purposes by the United States Military in the 1980s and early 1990s. This possibility was suggested to us by staff at the Ministry of Natural Resources who were contacted to provide clarification regarding seemingly erroneous coordinates given for certain Alvar sites. In such cases, as with many other aspects of the project, data ground truthing would be a vital improvement to the reliability of this project's deliverables.